Update to 8gnc — Brand Growth Diagnostic

{
  "name": "brandprint-tier-b",
  "description": "Use when the user asks for Tier-B brand elements — brand activator, brand platform word, tone and manner guidelines, moat statement, brand no-no's, brand guardrails, brand voice, or brand behavior testing — Layer 3 of the Brandprint research stack. Also trigger on \"run Tier-B,\" \"brand activation,\" \"brand tone,\" \"competitive moat,\" \"brand rules,\" or when the user wants practical, testable brand outputs derived from existing strategy research. Use immediately when the brandprint-tier-a skill has just completed — chain directly using its output as the seed.",
  "included_files": [
    {
      "relative_path": "references/output-schema-tier-b.md",
      "size_in_bytes": 3034
    }
  ],
  "skill_md_contents": "---\nname: brandprint-tier-b\ndescription: Use when the user asks for Tier-B brand elements — brand activator, brand platform word, tone and manner guidelines, moat statement, brand no-no's, brand guardrails, brand voice, or brand behavior testing — Layer 3 of the Brandprint research stack. Also trigger on \"run Tier-B,\" \"brand activation,\" \"brand tone,\" \"competitive moat,\" \"brand rules,\" or when the user wants practical, testable brand outputs derived from existing strategy research. Use immediately when the brandprint-tier-a skill has just completed — chain directly using its output as the seed.\n---\n\n# Brandprint Tier-B — Activation & Guardrails Directive\n\nDeliver 5 Tier-B brand elements that are practical, testable, and aligned to validated Tier-A truths. Evidence is lightweight but real. Every element gets at least one behavioral proxy signal.\n\n## When NOT to Use\n\n- **No Tier-A output exists.** Tier-B derives, it doesn't originate. Run `brandprint-tier-a` first — standalone mode here produces reduced-confidence output and says so in the audit log.\n- **Full visual identity or design systems.** Tone, platform word, and guardrails are verbal/behavioral elements. Logos, palettes, and type systems are design work outside this chain.\n- **Testing positioning against competitors.** Tier-B checks alignment with Tier-A, not differentiation in market — that's `competitive-positioning-audit` (Layer 5).\n\n## Chain Position\n\nThis is **Layer 3** of a 6-layer Brandprint research stack:\n\n1. **Core Human Truth** (Layer 1) → foundational truth sentence, tension map, archetypes, lexicon\n2. **Brandprint Tier-A** (Layer 2) → 10 defensible brand strategy elements with evidence-gated sprints\n3. **Brandprint Tier-B** (this skill) → uses Layer 2 output as seed; produces 5 actionable brand elements with proxy tests\n4. **Brandprint Tier-C** (Layer 4) → uses Tier-A + Tier-B output as seeds; produces 4 stylistic elements ready for deployment\n5. **Competitive Positioning Audit** (Layer 5) → validates differentiation against named competitors; may trigger second-pass refinement\n6. **Brand Strategy Report** (Layer 6) → compiles all layers into consulting-grade deliverable\n\n**When Layer 2 has just completed:** Import its full output JSON directly. Pull Tier-A anchors — tensions, outcomes, audience, equity ladder, competition, tagline, moat signals — into the Intake phase. Do not re-research what Layers 1–2 already validated. Carry forward all sources and the contradiction matrix.\n\n**When running standalone:** Resolve all variables with the user. If no Tier-A backbone exists, build a lightweight Backbone Tap from scratch using desk research, but note reduced confidence in the audit log.\n\n## Variables to Resolve\n\nBefore starting, confirm these with the user (or inherit from Layers 1–2):\n\n| Variable | What to Capture | Default |\n|----------|----------------|---------|\n| `brand_name` | The brand being researched | *required* |\n| `topic` | Product, service, or subject | *required* |\n| `category` | Industry or category | *required* |\n| `audience` | Primary audience definition | *required* (or from Layer 2) |\n| `region_context` | Geography or cultural context | US / English-speaking |\n| `constraints` | Scope, legal, brand safety, cultural sensitivities | None |\n| `languages` | Output language | English |\n| `timeline_days` | Suggested execution window | 7–21 days |\n| `mode` | `rapid`, `standard`, or `enterprise` | `standard` |\n| `backbone_source` | Link or reference to Tier-A backbone repository | From Layer 2 if chained |\n\n## Principles (Non-Negotiable)\n\n1. **Evidence-informed, not evidence-bloated.** Two independent sources or one source + a behavioral proxy per claim.\n2. **Mirror the audience's words.** Reduce adjectives. Show behaviors.\n3. **Test the smallest possible thing:** five-second comprehension, recall ping, micro-click, or completion rate.\n4. **If an output conflicts with Tier-A,** either revise or state boundary conditions.\n5. **No fabrication.** If data is unknown, label as unknown.\n\n## Evidence & Citation Policy\n\nFor any external claim, include: title, publisher, URL, dates, and a one-line evidence note. Quote sparingly. Same citation format as Layers 1–2.\n\n## Global Acceptance Gates\n\nEvery Tier-B element must meet these before its claim is accepted:\n\n- **Proof minimum:** ≥2 independent sources OR 1 source + 1 behavioral proxy\n- **Plain-language and confusion checks** for anything public-facing\n- **Alignment check** against Tier-A tensions, outcomes, and audience\n\n**Mode adjustments:**\n- *Rapid:* Single pass using Backbone + 1 proxy test per element\n- *Standard:* Backbone + 2 proxy tests per element; alt variants where helpful\n- *Enterprise:* Add cross-segment variants and pre-registered test plans\n\n## What NOT to Do\n\n- Don't invent survey results\n- Don't contradict Tier-A without stating limits\n- Don't ship jargon\n- Don't skip proxy tests — every element needs at least one behavioral signal\n- Don't copy Tier-A proof points verbatim as proxy tests — proxy tests must be designed fresh per element (recall ping, micro-click, completion rate)\n- Don't treat the platform word as a tagline — it is a one-word decision filter for internal use, never public copy\n- Don't write a no-no rule without its rationale and the Tier-A tension it protects — an unanchored rule gets ignored or argued away\n\n## Workflow\n\n### Phase 0: Intake & Alignment\n\n**Goal:** Confirm context and pull Tier-A anchors.\n\nActions:\n- Resolve all variables; pull Tier-A anchors: tensions, outcomes, audience, equity ladder, competition\n- List constraints (legal, brand safety, cultural sensitivities)\n- Define success criteria per element (see acceptance gates in each sprint)\n\nOutputs: `alignment_brief`, `tierA_anchor_summary`, `success_criteria`\n\n---\n\n### Phase: Backbone Tap\n\n**Goal:** Reuse existing evidence; no re-collection if unnecessary.\n\nActions:\n- Inherit review verbatims, search-intent phrases, competitor inventory from Tier-A Backbone\n- Flag contradictions to watch\n\nIf no Tier-A Backbone exists, conduct lightweight desk research to populate: review verbatims (min 20), search-intent sample (min 10 phrases), top 5 competitor inventory. Log reduced confidence.\n\nOutputs: `backbone_refs_used`, `contradictions_watchlist`\n\n---\n\n### Sprint: Brand Activator (signature micro-behavior)\n\n**Goal:** Define the one repeatable action or ritual that makes the brand tangible at touchpoints.\n\nActions:\n- Hypothesize 2–3 activators derived from Tier-A tensions and outcomes\n- Define where to use (touchpoints) and expected micro-behavior\n- Design a proxy test: completion rate, post-exposure recall, or micro-conversion\n- Select winner based on proxy signal + feasibility\n\n**Acceptance gates:**\n- Proof minimum: ≥2 (independent sources or 1 source + 1 behavioral proxy)\n- Recall or completion proxy required\n- Alignment with Tier-A required\n\nOutputs: `activator_candidates`, `test_plan`, `proxy_results`, `final_activator_pack`\n\n---\n\n### Sprint: Brand Platform (single word/idea)\n\n**Goal:** Identify one word that becomes the decision filter for the brand.\n\nActions:\n- Mine audience language to propose 3–5 platform words with synonyms\n- Run semantic field check against corpus and competitors\n- Write the gatekeeper question: \"Does this create more [word]?\"\n- Select word with strongest evidence and lowest confusion risk\n\n**Acceptance gates:**\n- Proof minimum: ≥2\n- Confusion rate check required\n- Semantic overlap with competitors must be noted\n\nOutputs: `platform_word_options`, `semantic_notes`, `gatekeeper_question`, `final_platform_word`\n\n---\n\n### Sprint: Tone & Manner (adjectives and key phrases)\n\n**Goal:** Codify how the brand sounds using real customer language.\n\nActions:\n- Extract frequent customer phrases; remove category clichés\n- Propose 3–6 adjectives and 5–8 key phrases grounded in verbatims\n- Run five-second and plain-language checks\n- Provide do/say examples and anti-patterns\n\n**Acceptance gates:**\n- Proof minimum: ≥2\n- Plain-language check required\n- Anti-patterns required\n\nOutputs: `adjectives`, `key_phrases`, `dos_and_donts`, `readability_notes`\n\n---\n\n### Sprint: Our Moat (why we win)\n\n**Goal:** Articulate why the brand wins in observable, defensible terms.\n\nActions:\n- Draft moat statements tied to observable advantages (speed to value, switching cost, IP, distribution)\n- Cross-check against competitor claims and public signals\n- Provide time-to-value comparison or switching-cost proxy\n\n**Acceptance gates:**\n- Proof minimum: ≥2\n- Observable signal required\n- No handwaving — every moat claim must cite a real signal\n\nOutputs: `moat_statement`, `evidence_notes`, `observable_signals`, `time_to_value_or_switch_cost_proxy`\n\n---\n\n### Sprint: Brand No-No's\n\n**Goal:** Define the guardrails that protect brand integrity.\n\nActions:\n- From complaints, usability findings, and brand safety constraints, list 5–10 \"never/avoid\" rules\n- Attach rationale and the risk each prevents\n- Map each to a Tier-A tension it protects\n\n**Acceptance gates:**\n- Proof minimum: ≥2\n- Each rule requires rationale\n- Link to Tier-A tension required\n\nOutputs: `no_nos_list`, `rationales`, `risk_mapping`\n\n---\n\n### Phase: Integration & Packaging\n\n**Goal:** Ensure Tier-B elements align and don't step on Tier-A.\n\nActions:\n- Run contradiction check against Tier-A anchors\n- Assemble quick rationale notes and citations\n- Produce a one-pager per element for handoff\n\nOutputs: `contradiction_check`, `rationale_notes`, `tierB_one_pagers`\n\n## Sprint Dependency Map\n\nTier-B sprints are lighter than Tier-A. Most can run in parallel after the Backbone Tap:\n\n```\nTier-A Output (Layer 2)\n    │\n    ▼\nIntake & Alignment\n    │\n    ▼\nBackbone Tap ──┬── Sprint: Brand Activator ──────────────┐\n               │                                          │\n               ├── Sprint: Brand Platform ────────────────┤\n               │                                          │\n               ├── Sprint: Tone & Manner ─────────────────┤\n               │                                          │\n               ├── Sprint: Our Moat ──────────────────────┤\n               │                                          │\n               └── Sprint: Brand No-No's ─────────────────┤\n                                                          │\n                         Integration & Packaging ◄────────┘\n```\n\nAll 5 sprints can run in parallel since they each tap the same Backbone + Tier-A anchors independently. The Integration phase waits for all 5 to complete.\n\n## Heuristics\n\n- If you can't explain it in one breath, it isn't Tier-B\n- Prefer real customer words over clever synonyms\n- One proxy signal beats five opinions\n- If two good options tie, pick the one that's cheaper to test live\n\n## Output Format\n\nDeliver the final output as structured JSON conforming to the schema in `references/output-schema-tier-b.md`.\n\nThe JSON must include: `brand_name`, `topic`, `audience`, `backbone_refs_used`, `tierB_results` (all 5 elements with proxy tests), `sources`, and `audit_log`.\n\n**Forward chaining:** When this skill completes, its full output object becomes the input seed for Layer 4 (Brandprint Tier-C). Preserve the complete JSON — Layer 4 needs Tier-B tone, platform word, and no-no rules. Also preserve the Tier-A JSON from Layer 2; Layer 4 needs both. If the `brandprint-tier-c` skill is available, proceed directly into it without user confirmation.\n\nIf the user has the `branded-mayhem-pdf` skill available, offer to generate a branded PDF deliverable of the Tier-B results.\n\n## Evaluation Rubric\n\n1. **Alignment** — Tier-B outputs reflect Tier-A truths without contradiction\n2. **Evidence** — each element meets the Tier-B proof minimum (≥2 or 1+proxy)\n3. **Behavior** — at least one proxy signal per element in rapid mode; two in standard mode\n4. **Clarity** — plain-language and low confusion in public-facing pieces\n5. **Differentiation** — platform word and moat are not commodity language\n6. **Utility** — activator is specific, placeable, and testable at live touchpoints\n\n## File I/O Contract (orchestrated mode)\n\nWhen an authorized orchestrator provides explicit paths, honor them exactly:\n\n- **Seed inputs:** read ONLY the JSON/YAML files listed in the dispatch prompt.\n- **Output:** write the final JSON object to the exact path given (under\n  `.brandprint/engagements/{slug}/pass-N/`). No other location.\n- **Return value:** your final message is the output path plus the layer's key\n  artifact — not the full JSON. The orchestrator reads files, not transcripts.\n\nWhen invoked as a direct skill call, present the JSON in conversation or through the supported file workflow and tell the user to preserve it for the next layer.\n"
}